Quantcast
Channel: Embedded Community : All Content - All Communities
Viewing all 3032 articles
Browse latest View live

Roving Reporter: Boosting Communications Performance with Embedded SSDs

$
0
0

To handle today’s massive and ever-increasing amounts of data traffic, business communications equipment of all types (commercial, corporate, small-and-medium business, and military/aeronautics) require performance. Every system component, large and small, must be meticulously selected to eke out greater performance gains.

 

One performance-enhancing solution with a big impact for its size is the embedded solid-state drive (SSD). Unlike SSDs packaged to replace hard disk drives (HDDs), embedded SSDs are tiny, inexpensive chips with capacity for an amazing amount of gigabytes. Soldered directly onto a board, they save space and increase ruggedness. Mounted on a disk on module (DOM) or an M.2 card, embedded SSDs give communications equipment manufacturers flexibility, and scalability in system configuration (Figure 1).

 

form factors.JPG

Figure 1. Mounted on a disk on module (DOM) or an M.2 card, embedded SSDs give communications equipment manufacturers flexibility and scalability in system configuration. 

 

Embedded SSD Advantages

The obvious reason to use embedded SSDs is they are faster and superior to HDDs. Where HDDs may require more than a minute to boot up a server, an SSD can do it in seconds. With their fast NAND flash technology, SSDs excel in performing random read/write operations, including retrieving all the small files read during the boot sequence.

 

Embedded SSDs also enhance virtual memory mapping performance. When memory is full and a server assigns memory functions to the boot drive, an HDD’s slow read/write speeds bog down the system. Innately faster SSDs deliver better virtual memory performance.

 

Other advantages drive interest in embedded SDDs as well. With no moving parts, SDDs can tolerate environments with higher ambient temperatures and more vibration. They save precious space in small form factors. They consume much less power than an HDD. And recent technology breakthroughs make them competitive in price.

 

Server-Class Embedded SDDs

One source for manufacturers looking for server-class embedded SDDs is Silicon Motion. Their FerriSSD* family of products include the SM659 and SM619 SATA single-package 6Gb/s SSDs. These compact solutions integrate Silicon Motion's NAND flash controller, industry-standard NAND flash memory, and DRAM in a small 90-ball, 1.0mm pitch BGA package (16x20x1.8mm).

 

The DRAM is key. With its low latency and tolerance for write-heavy applications, DRAM enables these SDDs to deliver best-in-class performance of up to 80,000 random input/output operations per second (IOPs). DRAM also increases reliability and extends SSD life.

 

The SM659 and SM619 feature a number of advanced Silicon Motion technologies and capabilities:

  • Advanced NAND Flash management, including error correction, bad block management, cell health monitoring, and automatic recovery
  • IntelligentScan with DataRefresh for enhanced data retention and read-disturbance protection
  • PowerShield for advanced protection against sudden power loss
  • DataPhoenix for instant data recovery
  • Global wear-leveling for evenly distributing program/erase cycles across all NAND cells and a low-write amplification index for maximizing SSD lifespan
  • Rigorous factory testing for ensuring extremely low defect rates and failure rates through product usage life (Figure 2)

 

erase cycle life.JPG

Figure 2. FerriSSDs undergo rigorous testing at customer-specified temperature to extremely low defect rates and failure rates through product usage life.

 

Silicon Motion recommends the SM659 for high reliability applications that require up to 30K Program/Erase (P/E) cycles and the SM619 for cost-effective systems with less stringent endurance requirements. FerriSSDs are available in commercial (0°C to 70°C) and industrial temperature grades (-40°C to 85°C).

 

FerriSSDs in Action

The SM659 and SM619 can play a number of roles in blade servers, application servers, routers, telecom base stations, network firewall servers, VoIP servers, and business telecom systems. They can function as a boot drive, embedded system storage element, backup recovery device, license drive, or storage in rugged communications equipment deployments where mechanical HDDs would fail.

 

Integrating FerriSSDs in System Designs

The SM659 and SM619 are typically installed as SSD modules in SATA DOM or M.2 form factors or installed directly onboard. The BGA-SDD form factor enables FerriSSDs to be placed virtually anywhere inside a tight server chassis, including a micro server’s extremely restricted space. FerriSSDs can be soldered down directly to the main PCB or an OEM’s proprietary module form factor.

 

FerriSSDs’ high reliability matches well with the high reliability and performance of solutions based on Intel® Xeon® processors. A particularly good fit for many of the communications equipment systems mentioned here is the Intel® Xeon® processor D product family. This product family brings the performance and advanced intelligence of  Intel Xeon processors into a dense, power-efficient system-on-a-chip (SoC) – see Figure 3. 

 

Xeon D.png

Figure 3. The Intel® Xeon® processor D product family brings the performance and advanced intelligence of Intel® Xeon® processors into a dense, power-efficient system-on-a-chip (SoC)

 

Based on industry-leading 14 nm process technology, the Intel Xeon processor D product family includes integrated platform controller hub (PCH) technology, two integrated 10 Gigabit Intel® Ethernet ports, and thermal design power (TDP) ranging from 20 W to 65 W all in a BGA package enabling rugged designs.  . These processors run the same instruction set as other Intel Xeon processor families, providing software consistency from the data center to the mobile edge.

 

Available with up to 16 cores, the SoCs offer exceptional node performance, up to 24 MB of last-level cache (LLC), and high-speed DDR4 memory support. Intel® Turbo Boost Technology 2.0 can dynamically increase processor frequency to deliver an extra burst of speed when needed and increased energy efficiency for less demanding tasks.

In addition to these capabilities, the Intel® Xeon® processor D-1500 product family includes the following advanced server-class capabilities:

  • Enhanced reliability, availability, and serviceability (RAS) features that include support for ECC memory and platform-level error management and resilience
  • Intel® QuickData technology for offloading memory accesses to the SoC for fast data movement with low processor overhead
  • Intel® Platform Storage Extensions to enable smarter and more cost-effective storage solutions that accelerate data movement, protect data, and simplify data management
  • Intel® Trusted Execution Technology (Intel® TXT) for platform verification that strengthens security while reducing performance impact
  • Intel® Advanced Encryption Standard New Instructions (Intel® AES-NI) for accelerating data encryption and decryption

 

Speed Up Design by Starting at the Solutions Directory

The Intel® Internet of Things Solutions Alliance’s Solution Directory gives OEMs and designers a head start. A variety of boards and systems are available with the Intel Xeon processor D product family, many with appropriate connectors for DOM or M.2 modules. Alliance members also offer custom design services for OEMs and designers wishing to purchase boards with the FerriSSD already soldered in place.

 

 

Learn More

Contact featured member:

 

Solutions in this blog:

 

Related topics:

Silicon Motion is a General member of the Intel® Internet of Things Solutions Alliance.

 

Mark Scantlebury

Roving Reporter (Intel Contractor), Intel® Internet of Things Solutions Alliance

Editor-in-Chief,Embedded Innovator magazine


Roving Reporter: Empowering vCPE Services with Layer 7 Visibility

$
0
0

Virtual Customer Premise Equipment (vCPE) provides an opportunity for service providers to offer new value-added services. Examples include next generation firewalls, application-specific QoS, or service chaining with a web-like delivery model and a high degree of personalization. To enable these opportunities, vCPE solutions need to offer visibility into application-level network activity – that is, Layer 7 visibility.

 

Qosmos, a leader in IP traffic classification and network intelligence technology for physical, SDN and NFV architectures, has the solution. The company offers a Layer 7 classification engine designed to enable vCPE solutions to utilize real-time application and subscriber information (Figure 1). In this post we explore the features and benefits of the Qosmos ixEngine. We then look at the optimal hardware engine and open software for delivering the required performance in enhanced packet processing and forwarding capabilities.

 

schema_xengine-hd.jpg

Figure 1. The Oosmos ixEngine is a Layer 7 classification engine designed to enable vCPE solutions to use real-time application and subscriber information.

 

The Opportunity and the Challenge

vCPE is a lead application for Network Function Virtualization (NFV). It provides an alternative way of delivering broadband services where most of the CPE functions are delivered by the service provider’s network and located near the service edge.

 

With vCPE, service providers can simplify CPE and increase service agility by hosting all virtualized CPE functionality in the network at a point of presence (PoP) or in another type of data center (Figure 2). Services such as DHCP, firewall, NAT, routing, VPN, and more, are delivered by virtual network functions (VNFs) running on generic, high-volume virtual machine (VM) instances configured for each broadband subscriber.

 

Qosmos_vCPE-diagram-web.jpg

Figure 2. With vCPE, service providers can simplify CPE and increase service agility by hosting all virtualized CPE functionality in the network at a point of presence (PoP) or in another type of data center.

 

vCPE delivers capex and opex savings because it doesn’t need proprietary hardware and reduces service truck rolls to remote offices by operating at the PoP. It also enables an app store model for VNFs. Operators can create a catalog of software-based services that can be deployed on demand using self-service portals.

 

To offer such tailored vCPE services through service function chaining (SFC) and optimize use of bandwidth and computing resources, vCPE solutions need embedded service classification based on Layer 7 information. However, vCPE solution providers may face restrictions in their product offerings due to traffic visibility limited to Layers 1-4. The value of firewalls, QoS, service chaining, and reporting could be greatly improved by leveraging the complete spectrum of Layer 1 to 7 information. This information is key in understanding which application is generating which flow on the network and applying the right service chain.

 

The Qosmos Layer 7 Classification Engine

Qosmos treats the network as a real-time database, and is able to identify, query and extract specific data with unparalleled precision and detail. The Qosmos ixEngine is a software development kit (SDK) that uses deep packet inspection (DPI) to provide IP classification and metadata extraction up to Layer 7 based on real-time application and subscriber information. It is easily integrated into vCPE solutions to offer stronger security, QoS, and reporting.

 

While some technologies are limited to identifying the application behind an IP flow, the Qosmos ixEngine goes further. It also extracts protocol and application metadata. This metadata enables developers to inject application-level insight into their solutions for complete visibility into network traffic in real time and a detailed understanding of network transactions and user behavior. Metadata extraction includes volume, application usage, application performance, identifiers, content, and file metadata. The Qosmos ixEngine also provides extension modules for aggregated and computed metadata.

 

Advantages of the Qosmos solution include the following:

  • Delivers high recognition rate: ability to identify all layers from Layer 2 to 7 in the OSI model
  • Includes 2500 classified and continuously updated protocols and 4300 application metadata extracted
  • Identifies protocols and applications based on flow pattern matching, session correlation, heuristics, and statistical analysis
  • Provides a modular architecture (flow management, regular expression engine, http parsing, etc.)
  • Allows users to develop their own signature plugins
  • Enables up to 10 Gbps (depending on traffic patterns and networking environment) per core on Intel® processors

 

Layer 7 visibility can be deployed in vSwitch, service functions, or VNFs, to perform traffic classification and metadata extraction. For future proofing, Qosmos ixEngine can also be configured using reference implementations such as OpenDaylight SFC.

 

Designed by Qosmos with developers in mind, the Qosmos ixEngine accelerates product development cycles. Its ready-to-use software libraries make it easy to embed IP classification and metadata extraction information into existing solutions, plus offer an additional toolkit for developing customized protocol plugins. The solution includes fully documented APIs, tutorials, and a large array of code samples and reference designs to facilitate integration into your solution.

 

Optimized for Intel® Xeon® Processors

The Qosmos ixEngine is optimized for Intel® technology to deliver the performance required for vCPE applications. The Qosmos ixEngine features built-in multi-core support capabilities such as optimized multi-thread support for scalability up to 96 cores and optimized code for high performance multicore processors and hardware acceleration.

 

To avoid latency from the Linux kernel when extracting metadata and content from packets flowing through the network, Qosmos uses the Data Plane Development Kit (DPDK). This set of software libraries and drivers developed by Intel is now available as open source software. DPDK provides enhanced packet processing and forwarding capabilities, enabling Intel® Xeon® processor-based servers to deliver very high packet throughput rates.

 

Suppliers of equipment, platforms, middleware, and software can use this Qosmos and Intel technology synergy to rapidly build application-aware solutions for service providers. The Qosmos ixEngine is especially well suited to enable intelligent, dynamic service chaining for cloud-based vCPE environments (Figure 3).

 

ref arch.JPG

Figure 3. Reference architecture for an L7-based service chaining solution based on the Qosmos ixEngine.

 

Advantages of the Latest Intel® Xeon® Processors

An ideal processor family for running the Qosmos ixEngine is the Intel® Xeon® processor D product family. With up to 16 cores, this advanced processor brings the performance and advanced intelligence of Intel Xeon processors into a dense, low-power-consumption system on a chip (SoC).

 

With enhanced reliability, availability, and serviceability features; platform storage extensions; and built-in hardware virtualization; the Intel Xeon processor D-1500 product family offers new options for optimizing a variety of communications workloads. It runs the same instruction set as the most powerful Intel Xeon processors to provide software consistency from the data center to the edge.

 

Make Layer 7 Visibility a Competitive Advantage

Learn more about the Qosmos ixEngine and how it delivers Layer 7 visibility with market-leading IP flow parsing technology to accelerate the delivery of application-aware solutions. And visit the Solutions Directory for a selection of boards and systems featuring the latest Intel Xeon processors.

 

 

Learn More

Contact featured member:

 

Solutions in this blog:

 

Related topics:

Qosmos is an Affiliate member of the Intel® Internet of Things Solutions Alliance

 

Mark Scantlebury

Roving Reporter (Intel Contractor), Intel® Internet of Things Solutions Alliance

Editor-in-Chief,Embedded Innovator magazine

eepromARMtool

$
0
0

I'm look for the eepromARMtool source, I need to program my new hardware with i210 on an ARM plat , please send the tool and related data to my E-mail :gongshengli@hnlx.com.cn (Already registe the EDC).

Thanks,

Gong

White Paper: Intel Xeon Cores Power High Performance and Low SWaP Complex Sensor Solutions for Aerospace & Defense Applications

$
0
0

Modern day processors for imaging, radar and other sensors are complex systems each unique with diverse processing requirements per company, platform, sensor, and application. They might include multiple sensor arrays each steerable and requiring multiple processes to control, point, display, and perform the data processing for a variety of image processing applications such as detection, recognition, and single target track as well as tracking multiple targets in the operational area.

 

A common feature is that these systems require high bandwidth and high performance signal processing. Such is now readily available with advances in FPGA capabilities as well as their programmability, the incorporation of embedded graphical processors, the increasing number of cores with vector units, and the high bandwidth of PCI Express and multi-Gigabit Ethernet.

Another common feature is that as the flexibility and power of these sensor systems has appreciated, so has the complexity of their software control. It is not uncommon for the architecture of embedded imaging and other sensor systems to have a framework of multiple software executables for the interface, data, and control processing.

 

For example, a passive infrared system in previous times might include a single monolithic imager which provided video from whatever was in front of where it was mounted. This simple technology-limited system didn’t require much sensor, line of sight (LOS), targeting, or display functionality to control the hardware or process the target information from past signal processing capabilities.

Included among the control processes are the user interface, system control, pointing and stabilization, sensor control, instrumentation, calibration, image display, mode control, single and multiple target track, etc. There might be a multiple instances of some of these targeting processes depending on the application.

 

Some of the processing on the control side in today’s software-driven systems has increased with higher frame rates, reduced latencies, more complex algorithms, and additional functionality. Some of the control processing has not increased, but engineers are not inclined to combine these software processes to avoid system development complexity. High-speed interfaces, FPGA, graphic, and multi-core digital signal processors have advanced sensor signal processing capabilities. The question is what can be done for the control side of the sensor processing.

 

This white paper will discuss how modern day processors can address the challenges of controlling today’s more complex sensor systems using an IR imaging system example. Download the paper.

White Paper: (Aerospace & Defense Applications) Deployable Core Counts Explode with Intel’s New Xeon D Processor

$
0
0

There are but a handful of silicon devices that have come along during the last several decades that could be considered “game changers” for embedded Defense applications. One such device was the single core PowerPC 7400, released by Motorola Semiconductor (formerly Freescale) and now NXP. The PPC 7400 included a specialized adjunct processor, the AltiVec SIMD (Single Instruction Multiple Data) Vector Engine. With AltiVec, DSP and linear algebraic functions like FFTs suddenly saw a huge performance leap, boosting execution by several orders of magnitude. The radar algorithms and EW techniques that relied on these mathematical constructs in turn ran faster by orders of magnitude. Defense programs were soon equipped with sensing, imaging and tracking capabilities that had not been possible before.

 

Fast forward to today; Intel has launched a new device that contains up to 16 cores. Each core contains a SIMD Vector Engine, known as the AVX2, which has twice the vector width of the original AltiVec engine. Core counts that once required multiple VME or VPX modules to realize are now embodied within a single device. The white paper examines this ground-breaking chip, the Xeon D processor, and explores some of the visionary defense applications that will now become practical.

Learn more about:

  • Genesis of the Xeon D
  • A Game-changer for Embedded Defense
  • Application Advantages
  • Enabling New Application Capabilities

 

Download the white paper

White Paper: Get the Big Embedded Defense Payoff - the 5th Gen Intel Core i7 Processor

$
0
0

Intel’s 5th Gen Core i7 "Broadwell" processor was developed for broad commercial markets but also offers big benefits to embedded defense applications. Tech refresh programs will see impressive performance-per-watt improvements while new designs can exploit an extremely fast multi-core processor and powerful integrated graphics combined in a single package.

 

Moving Forward with Intel’s ‘Tick-Tock’

The new 5th Generation Intel Core i7 ‘Broadwell’ processor is the latest step in Intel’s relentless march of performance improvement. Following a Tick-Tock model, Intel alternates its processor development by creating a new micro-architecture in one generation and then shrinking the die geometry in the following generation. This Tick- Tock cadence steps forward roughly every 18 months. Broadwell is a die shrink of the previous generation Haswell architecture chip.

 

Broadwell’s Extra Punch is the Integrated Graphics Processing Unit

For general purpose processing, Broadwell delivers a modest 10-15% improvement over the 4th generation Core i7 or a more significant 40-50% improvement over the 3rd Generation Core i7 family. However, a much larger payoff is evident in the chip’s graphics and floating point processing, driven by Broadwell’s integrated Graphics Processing Unit (GPU).

 

Download the white paper to learn more about:

  • Meeting a tech refresh size, weight and power (SWaP) challenge
  • Tech refreshes driven by application enhancements
  • Deployable SBCs
  • Intel processors for embedded defense applications

Webinar: The Quickening- Hardware acceleration speeds video processing through at the network edge

$
0
0

OTT video means that TV is now being consumed anywhere and real-time transcoding helps makes streaming as efficient as possible. At the same time, operators are leveraging software defined network (SDN) and network functions virtualization (NFV) technologies. Intelligence and capability is being pushed to the edge of the network, closer to users. In this new ‘edge revolution’, service providers need higher density, scalable computing power while lowering CapEx and OpEx.  The presentation will include application use cases and a description of a virtual network function demonstration that showed a greater than 20X performance improvement using open source and open standard software and hardware.

 

Hosted By: Artesyn Embedded Technologies

 

23rd August 2016

9AM PST | 12AM EST | 5PM BST

 

Speakers:
Daniel Leih - Product Marketing Manager, Acceleration Products, Artesyn
Abe Nejad - Anchor, TIA Now
Nagesh Puppala - Media Segment Director, Data Center Group, Intel
Guy Daniels - Director of Content, TelecomTV

 

View the webinar here (on demand):   Webinar Registration

如何下載Broadwell DE microcode

$
0
0

HI 大家好,

     我目前在開发coreboot Broadwell DE 的案子, .Config 中提及如下檔案, 請問這要去哪裡下載? 感謝.

../intel/cpu/broadwell_de/microcode/M1050663_07000001.h ../intel/cpu/broadwell_de/microcode/M1050662_0000000A.h ../intel/cpu/broadwell_de/microcode/MFF50661_F1000008.h

 

 

中國台灣

James


如果举办intel嵌入式学习活动,您对以下哪个活动最感兴趣?

How to down load broadwell DE micro code

$
0
0

Dear All,

     I porting coreboot Broadwell DE project, the .Config  need include below files, have some one can tell me goto where to download it? before Intel FSP have include microcode,

but broadwell and skylake look like no follow this rule, thanks.

../intel/cpu/broadwell_de/microcode/M1050663_07000001.h ../intel/cpu/broadwell_de/microcode/M1050662_0000000A.h ../intel/cpu/broadwell_de/microcode/MFF50661_F1000008.h

 

  Best Regards

 

-James

intel 210 nic flash-less mode

$
0
0

we want to use imx6 + 210(igb) implment Gb network, but the 210 board owns a flash chip. what's context in it? how to del it ? how to configure ? please help, thanks

Problem with second Eth controller (WG82574L)

$
0
0

Hi.

 

My design consists of an Q7 module based Intel ATOM processor, chipset US15W. On the carrier board of this module there are two ethernet ports, each controlled by an Intel WG82574L controller. Used OS is Windows XP embedded.

 

My problem is that I can't get one of the ethernet ports to work properly. See attached screen dump over the Win XP device manager. HW wise the two ports are identical, however there is always the same port that fails on all 5+ boards assembled.

 

I'm running out of ideas. The driver has been updated, both controllers are enabled. Moreover, since both controllers are detected by Win XP some kind of connection must be established.

 

My main question is, being a HW designer and not totally wizardous on the SW parts:

What can I consider to be working/not working when this exclamation mark, "!", is presented in the device manager?

 

Obiously, some kind of connection via PCEe must be working, right?

 

I would be very grateful for any idea or possible solution.

 

Thanks!

/Henrik

Skylake Power-efficient, Palm-sized Embedded System

$
0
0

Axiomtekt eBOX560-500-FLis an Intel® Skylake-based fanless embedded system with compact size and full-featured I/O interface. To bring both excellent performance and ultra-low power consumption, the palm-sized embedded box PC is powered by the 6th generation Intel® CoreTM i7-6600U or Celeron® 3955U processor (formally codename: Skylake). One 260-pin DDR4-2133 SO-DIMM socket with system memory up to 16 GB is available. The power efficient IP40 embedded system supports 12V DC input with screw-lock and features a user-friendly AT/ATX DIP switch. The compact eBOX560-500-FL is suitable for Industrial Internet of Things (IIoT) solutions, digital signage, retail equipment, smart factory automation, thin clients, industrial controller systems and many more.

 

The Axiomtek extremely compact eBOX560-500-FL has wide operating temperature ranges from -10°C to +50°C (-14°F to +122°F).  It features IP40-rated anti-dust ingress protection, and is able to endure vibration up to 3G. The reliable fanless embedded box computer comes with full-featured I/O interface and one PCI Express Mini Card for wireless network connection. The embedded controller is definitely one of the best solutions for IIoT applications.

 

The rugged industrial embedded computer, eBOX560-500-FL, is equipped with rich I/O connectivity including four USB 3.0 ports, one RS-232/422/485 port, one RS-232 port, two Gigabit Ethernet ports, two SMA type connector openings for antenna, and two HDMI ports. One 2.5” SATA HDD and one mSATA are available for storage. The space-saving fanless embedded platform supports Jumbo Frame (9.5K), wake-on-LAN (WoL), PXE Remote Boot and Teaming. It runs well with the Windows® 8.1, Windows® 10, and supports Axiomtek’s exclusive “AXView 2.0” monitoring software package. To fulfill various application needs, the eBOX560-500-FL supports VESA mount, wall mount and DIN-rail mount by optional requests.

 

Advanced Features:

  • 6th generation Intel® Core™ i7-6600U 3.4 GHz/ Celeron® 3955U 2.0 GHz Skylake ULT SoC
  • 260-pin DDR4-2133 SO-DIMM max. up to 16 GB
  • Compact size with 12VDC power input
  • Supports 2 HDMI, 2 COM and 4 USB 3.0 ports
  • Supports Jumbo Frame (9.5K),WoL, PXE and Teaming

eBOX560-500-FL Box PC

White Paper: Data Transport for OpenVPX HPEC in Defense & Aerospace Applications

$
0
0

The Perry Memo transformed the procurement of electronic computer equipment for the Department of Defense (DoD) in 1994 by creating the concept of Commercial Off-The-Shelf (COTS). This memo helped mold the COTS VME/VPX industry.  If we fast forward to today, the introduction of OpenVPX builds on the reality of an open standard eco-system for the development of next generation rugged computer systems.

 

The Data Transport for OpenVPX HPEC white paper introduces the basics of both the hardware and data transfer mechanisms for embedded HPEC systems. Curtiss-Wright will discusses the fabric, middleware, and backplane technologies required for high-speed data transfer in OpenVPX platforms.

 

Processing Hardware

Intel has been a leader in data processing by providing the required performance needed for HPEC based systems. Their memory speeds, cache sizes and vector processing are best-in-class, and Intel’s Internet of Things (IoT) group provides the support and services to maintain our success with their guaranteed seven-year part availability.

 

Producing processors in industrial grade temperature ranges and packaging them as a Ball Grid Array (BGA) device directly from the foundry provides significant advantages as well for the rugged, harsh environments of the embedded space.

 

Accelerators also play a significant role in HPEC class systems when it comes to processing because they are capable of performing the “heavy lifting” allowing the processor to assume the role of data traffic (I/O) manager. This way of processing/accelerators is proven in HPC systems with the latest supercomputers which deploy more GPUs than CPUs. The invention of deep learning, where GPU cores mimic neurons of the human brain, is another reason for the huge uptake in the acceptance of GPUs as accelerators in the last couple of years.

 

Download Curtiss-Wright'sData Transport for OPENVPX HPEC White Paper to learn more about:

  • OpenVPX
  • CPUs, GPUs, FPGAs
  • Fabric40
  • Control Plane
  • Data Plane
  • Expansion Plane
  • Mellanox Connect-3
  • PCI Express Switching
  • Remote Direct Memory Access

Ultra-Small Form Factor Mission Systems: Great Things Come in Small Packages

$
0
0

Aerospace and defense systems integrators continue to push for reductions in size, weight,power, and cost (SWaP-C) to support advanced sensor/vetronics payloads onboard manned and unmanned platforms. Fortunately, groundbreaking miniaturization of mission processor and network switch subsystems are enabling UAS (unmanned air system), UGV (unmanned ground vehicle), UUV (unmanned undersea vehicle), USV (unmanned surface vehicle), as well as manned fighter aircraft, helicopter, and tactical ground vehicle platforms to expand their mission capabilities.

 

Fortuitously, COTS technology to shrink electronic subsystems is rolling out and enabling systems integrators to more effectively support technology insertion of advanced processor and network backbone architectures. Size and weight constrained platforms often require multiple computer processing elements and sensors as part of their vehicle electronics payloads that are ultimately interconnected to gather and share information. This need to fit more electronics into a limited space envelope is driving the necessity for smaller and smaller processor and network connectivity solutions.

Ultra small form factor

The Ultra-Small Form Factor Mission Systems white paper introduces a new class of highly capable rugged COTS mission systems known as ultra-small form factor (USFF) and highlight how these line replaceable units (LRUs) are helping to drive down cost, capability tradeoffs, and SWaP in size and weight-sensitive platforms.

 

Learn more about:


Start from scratch with atom

$
0
0

Hi guy,

 

I was posted in wrong branch... (processors)

 

For my project I would like use an intel atom X7-8700.

I start with nothing I need to build everything (choose RAM, Flash etc...).

I would make my own board (like raspberry, odroid, ...) based on intel atom.

 

Firtsly :

 

Page 338 of this doc :

http://www.intel.fr/content/dam/www/public/us/en/documents/datasheets/atom-z8000-datasheet-vol-1.pdf

I don't understand the center of the footprint there is no dimension or anything else for draw it.

 

Secondly :

 

I do not use BIOS but a bootloader saved at the begins of flash memory.

I think i do use the BLDK software but i haven't got the .bsf file.

Where it is ?

 

I found u-boot, can it make the bootloader for my board?

I will send an e-mail to the developer to know more about u-boot.

 

Do you now how to use another method to start from scratch with intel's CPU?

 

In advance,

Thank's you

lanconf32 shuts the ethernet link down

$
0
0

Hello everybody,

 

I'm trying to use the lanconf32 tool on a board equipped with a Rangeley C2000 Atom processor with embedded MAC and an external Marvell PHY (88E1112).

The iqvlinux.ko driver is a v.1.1.5.3. Inserting and removing it works well.

Our kernel is based on a CentOS 6.3 32bits.

 

The lanconf32 utility detects the ethernet interface, but as soon as I select one of the interface, the initialization procedure shuts it down (both LEDs are now off).

Following transmit or receive tests fails, as the lanconf32 utility reports "No link".

 

Am I missing anything in the procedure ?

Any hints ?

Thanks in advance.

Best regards,

Patrick Agrain

X5 e8000具备USB 3.0 otg么?

Cherry Trail Z5-Z8300的资料有吗?

请问下:i3-4005u跟J1900那个性能好?

Viewing all 3032 articles
Browse latest View live


<script src="https://jsc.adskeeper.com/r/s/rssing.com.1596347.js" async> </script>